To conduct a meta-analysis of clinical trials that examined the effect of music-supported therapy on stroke-induced motor dysfunction, comprehensive literature searches of PubMed, Embase and the Cochrane Library from their inception to April 2016 were performed. A total of 10 studies (13 analyses, 358 subjects) were included; all had acceptable quality according to PEDro scale score. The baseline differences between the two groups were confirmed to be comparable. Compared with the control group, the standardized mean difference of 9-Hole Peg Test was 0.28 (−0.01, 0.57), 0.64 (0.31, 0.97) in Box and Block Test, 0.47 (0.08, 0.87) in Arm Paresis Score and 0.35 (−0.04, 0.75) in Action Research Arm Test for upper-limb motor function, 0.11 (−0.24, 0.46) in Berg Balance Scale score, 0.09 (−0.36, 0.54) in Fugl-Meyer Assessment score, 0.30 (−0.15, 0.74) in Wolf Motor Function Test, 0.30 (−0.15, 0.74) in Wolf Motor Function time, 0.65 (0.14, 1.16) in Stride length and 0.62 (0.01, 1.24) in Gait Velocity for total motor function, and 1.75 (0.94, 2.56) in Frontal Assessment Battery score for executive function. There was evidence of a positive effect of music-supported therapy, supporting its use for the treatment of stroke-induced motor dysfunction. This study was registered at PRESPERO (CRD42016037106).